Is there a way to make all un-selected splines faded? in a bivariate plot!
The default is same as pre-selection.
Thanks
You'll have to do this manually -- JMP does not give an automatic handle to this. I use Spline Coef() and Spline Eval() to generate the spline points, then use a graphics script to draw the lines and add a handle to the legend box.
Spline Coef()
Spline Eval()
go to analyze > fit y by x > then input y and x variables in contextual
then follow the next photo
Do you have a set up like this example, where the Group by menu command sets up groups of data before using the Fit Flexible command?
If so, then I do mot know what you mean by selected or un-selected splines. There is no way to select one of the fitted groups.
I believe that it is a pretty common situation when looking at a large number of trendlines (perhaps you have 50 tools all running the same process and are comparing defectivity vs. tool) that you'd want to be able to highlight a specific trend. The natural way of doing this is by selecting the trend in the legend, but this only emphasizes the points, not the lines. I've created a custom script that does this and I believe this is what the OP is asking about. Below is a comparison of my solution vs. a standard Row Legend behavior on some fake data.
The second plot here (the default behavior) is mostly useless w/re to comparing anything about the trendlines
JMP provides all the tool necessary to do this -- Spline Eval, Spline Coef, and handlers to the TableBox object to re-draw the lines when the legend selection is updated, but it can be a bit of a chore to setup initially.
Looks amazing @ErraticAttack
I'd appreciate if you could share your script.
Thanks!
